1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
|
\name{kqr-class}
\docType{class}
\alias{kqr-class}
\alias{alpha,kqr-method}
\alias{cross,kqr-method}
\alias{error,kqr-method}
\alias{kcall,kqr-method}
\alias{kernelf,kqr-method}
\alias{kpar,kqr-method}
\alias{param,kqr-method}
\alias{alphaindex,kqr-method}
\alias{b,kqr-method}
\alias{xmatrix,kqr-method}
\alias{ymatrix,kqr-method}
\alias{scaling,kqr-method}
\title{Class "kqr"}
\description{The Kernel Quantile Regression object class}
\section{Objects from the Class}{
Objects can be created by calls of the form \code{new("kqr", ...)}.
or by calling the \code{kqr} function
}
\section{Slots}{
\describe{
\item{\code{kernelf}:}{Object of class \code{"kfunction"} contains
the kernel function used}
\item{\code{kpar}:}{Object of class \code{"list"} contains the
kernel parameter used }
\item{\code{coef}:}{Object of class \code{"ANY"} containing the model parameters}
\item{\code{param}:}{Object of class \code{"list"} contains the
cost parameter C and tau parameter used }
\item{\code{kcall}:}{Object of class \code{"list"} contains the used
function call }
\item{\code{terms}:}{Object of class \code{"ANY"} contains the
terms representation of the symbolic model used (when using a formula)}
\item{\code{xmatrix}:}{Object of class \code{"input"} containing
the data matrix used }
\item{\code{ymatrix}:}{Object of class \code{"output"} containing the
response matrix}
\item{\code{fitted}:}{Object of class \code{"output"} containing the
fitted values }
\item{\code{alpha}:}{Object of class \code{"listI"} containing the
computes alpha values }
\item{\code{b}:}{Object of class \code{"numeric"} containing the
offset of the model.}
\item{\code{scaling}}{Object of class \code{"ANY"} containing
the scaling coefficients of the data (when case \code{scaled = TRUE} is used).}
\item{\code{error}:}{Object of class \code{"numeric"} containing the
training error}
\item{\code{cross}:}{Object of class \code{"numeric"} containing the
cross validation error}
\item{\code{n.action}:}{Object of class \code{"ANY"} containing the
action performed in NA }
\item{\code{nclass}:}{Inherited from class \code{vm}, not used in kqr}
\item{\code{lev}:}{Inherited from class \code{vm}, not used in kqr}
\item{\code{type}:}{Inherited from class \code{vm}, not used in kqr}
}
}
\section{Methods}{
\describe{
\item{coef}{\code{signature(object = "kqr")}: returns the
coefficients (alpha) of the model}
\item{alpha}{\code{signature(object = "kqr")}: returns the alpha
vector (identical to \code{coef})}
\item{b}{\code{signature(object = "kqr")}: returns the offset beta
of the model.}
\item{cross}{\code{signature(object = "kqr")}: returns the cross
validation error }
\item{error}{\code{signature(object = "kqr")}: returns the
training error }
\item{fitted}{\code{signature(object = "vm")}: returns the fitted values }
\item{kcall}{\code{signature(object = "kqr")}: returns the call performed}
\item{kernelf}{\code{signature(object = "kqr")}: returns the
kernel function used}
\item{kpar}{\code{signature(object = "kqr")}: returns the kernel
parameter used}
\item{param}{\code{signature(object = "kqr")}: returns the
cost regularization parameter C and tau used}
\item{xmatrix}{\code{signature(object = "kqr")}: returns the
data matrix used}
\item{ymatrix}{\code{signature(object = "kqr")}: returns the
response matrix used}
\item{scaling}{\code{signature(object = "kqr")}: returns the
scaling coefficients of the data (when \code{scaled = TRUE} is used)}
}
}
\author{Alexandros Karatzoglou\cr \email{alexandros.karatzoglou@ci.tuwien.ac.at}}
\seealso{
\code{\link{kqr}},
\code{\link{vm-class}},
\code{\link{ksvm-class}}
}
\examples{
# create data
x <- sort(runif(300))
y <- sin(pi*x) + rnorm(300,0,sd=exp(sin(2*pi*x)))
# first calculate the median
qrm <- kqr(x, y, tau = 0.5, C=0.15)
# predict and plot
plot(x, y)
ytest <- predict(qrm, x)
lines(x, ytest, col="blue")
# calculate 0.9 quantile
qrm <- kqr(x, y, tau = 0.9, kernel = "rbfdot",
kpar = list(sigma = 10), C = 0.15)
ytest <- predict(qrm, x)
lines(x, ytest, col="red")
# print model coefficients and other information
coef(qrm)
b(qrm)
error(qrm)
kernelf(qrm)
}
\keyword{classes}
|
